Tencent Advertising Algorithm Challenge 2025: All-Modality Generative Recommendation

2026-04-04

AI Summary

This work addresses the absence of large-scale, real-world, and fully multimodal (text, image, etc.) public benchmarks for generative recommendation in industrial advertising scenarios. To bridge this gap, the authors introduce TencentGR-1M and TencentGR-10M—two novel multimodal generative recommendation datasets derived from anonymized real-world Tencent advertising logs. These datasets support user behavior sequence modeling and multimodal content generation, while explicitly distinguishing between click and high-value conversion behaviors through a weighted evaluation mechanism. The study employs state-of-the-art embedding models to extract multimodal representations and constructs baseline systems integrating collaborative IDs with autoregressive sequences. Furthermore, the team organized a global algorithm challenge and open-sourced both datasets and code, significantly advancing research and practical applications in industrial-scale multimodal generative recommendation.

Abstract

Generative recommender systems are rapidly emerging as a new paradigm for recommendation, where collaborative identifiers and/or multi-modal content are mapped into discrete token spaces and user behavior is modelled with autoregressive sequence models. Despite progress on multi-modal recommendation datasets, there is still a lack of public benchmarks that jointly offer large-scale, realistic and fully all-modality data designed specifically for generative recommendation (GR) in industrial advertising. To foster research in this direction, we organised the Tencent Advertising Algorithm Challenge 2025, a global competition built on top of two all-modality datasets for GR: TencentGR-1M and TencentGR-10M. Both datasets are constructed from real de-identified Tencent Ads logs and contain rich collaborative IDs and multi-modal representations extracted with state-of-the-art embedding models. The preliminary track (TencentGR-1M) provides 1 million user sequences with up to 100 interacted items each, where each interaction is labeled with exposure and click signals, while the final track (TencentGR-10M) scales this to 10 million users and explicitly distinguishes between click and conversion events at both the sequence and target level. This paper presents the task definition, data construction process, feature schema, baseline GR model, evaluation protocol, and key findings from top-ranked and award-winning solutions. Our datasets focus on multi-modal sequence generation in an advertising setting and introduce weighted evaluation for high-value conversion events. We release our datasets at https://huggingface.co/datasets/TAAC2025 and baseline implementations at https://github.com/TencentAdvertisingAlgorithmCompetition/baseline_2025 to enable future research on all-modality generative recommendation at an industrial scale. The official website is https://algo.qq.com/2025.

Problem

• generative recommendation
• all-modality
• advertising
• benchmark dataset
• multi-modal

Innovation

• Generative Recommendation
• All-Modality
• Multi-Modal Embedding
• Industrial-Scale Benchmark
• Conversion-Aware Evaluation